Mesoporous silica particles (Davisil) were functionalized with aminopropyltriethoxysilane (APTES) in a fractional factorial design with 19 different synthesis and uptake experiments. The number of amino groups and the uptake of CO(2) were optimized in a 2(V)(5-1) design. Most important to functionalizationwas the amount of water present during synthesis, the reaction time, and pretreating the silica with a mineral acid; certain two-way interactions were shown to be statistically significant as well. Modifications performed at 110 or 80 degrees C showed no significant differences concerning amine content or uptake of CO(2). Properly choosing center points for the discrete variables is problematic and is somewhat related to the lack of fit with respect to CO(2) uptake; the regression was good. Solid-state (29)Si NMR showed that the APTES was mainly fully condensed. Specific surface areas did not correlate with the number of n-propylamine groups on the silica, which is indicative of differential levels of heterogeneity in the coverage of propylamines. The uptake of CO(2) and N(2) was measured from -20 to 70 degrees C and from 0 to 1 bar and parametrized by the Freundlich isotherm. Amine-modified silica adsorbed significant amounts of CO(2), especially at the low partial pressure, which is important for CO(2) capture from flue gas. At such pressures, samples with a high density of amine (4 amines/nm(2)) showed a much higher uptake of CO(2) than did those with densities of similar to 2-3 amines/nm(2), reflecting differential tendencies to form propylammonium-propylcarbamate ion pairs; these require close proximity among amine groups to form. Water affected the uptake of carbon dioxide in different ways. Certain samples took up more moist CO(2) gas than dry CO(2), and others took up less moist CO(2) than dry CO(2), which is indicative of differential tendencies toward water adsorption. We conclude that experimental design is a time-efficient approach to the functionalization of silica with propylamine groups.